Addressee and Response Selection for Multilingual Conversation
This addresses the challenge of building multilingual conversational AI systems, particularly for low-resource languages, though it appears incremental.
The paper tackles multilingual addressee and response selection in conversational systems by introducing knowledge transfer methods to leverage high-resource language data for low-resource languages, and experiments on a new dataset show their effectiveness.
Developing conversational systems that can converse in many languages is an interesting challenge for natural language processing. In this paper, we introduce multilingual addressee and response selection. In this task, a conversational system predicts an appropriate addressee and response for an input message in multiple languages. A key to developing such multilingual responding systems is how to utilize high-resource language data to compensate for low-resource language data. We present several knowledge transfer methods for conversational systems. To evaluate our methods, we create a new multilingual conversation dataset. Experiments on the dataset demonstrate the effectiveness of our methods.